bs_bs_banner

Geriatr Gerontol Int 2016; 16: 57 7 –585

ORIGINAL ARTICLE: EPIDEMIOLOGY, CLINICAL PRACTICE AND HEALTH

Muscle strength: A better index of low physical performance than muscle mass in older adults Yeo Hyung Kim,1 Kwang-Il Kim,2 Nam-Jong Paik,1 Ki-Woong Kim,3,4 Hak Chul Jang2 and Jae-Young Lim1 Departments of 1Rehabilitation Medicine, 2Internal Medicine, 3Neuropsychiatry, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, and 4Department of Brain and Cognitive Science, Seoul National University College of Natural Sciences, Seoul, Korea

Aim: The most appropriate muscle index for the definition of sarcopenia has not been agreed on. We aimed to investigate the associations of muscle mass and strength with 5-year mortality and low physical performance. Methods: We included 560 participants aged 65 years or older in the analysis. Muscle and fat mass were assessed by dual-energy X-ray absorptiometry. Leg muscle and grip strength were measured using dynamometers. The clinical outcomes were 5-year mortality and low physical performance (Short Physical Performance Battery score < 9) in 5 years. Associations between muscle indices and clinical outcomes were analyzed. A Cox proportional hazard model for mortality and a logistic regression model for physical performance were used. Results: Decreases in leg muscle and grip strength were significantly associated with 5-year mortality and low physical performance in both sexes. Total muscle mass in men and appendicular skeletal mass in both sexes were associated with mortality, but not with low physical performance. Lower leg muscle strength (OR 0.107; P = 0.020) was an independent predictor of low physical performance in women after adjusting for age, fat, cognition, and depression. Lower leg muscle (OR 0.123; P = 0.031) and lower grip strength (OR 0.950; P = 0.012) were independent predictors of low physical performance in men. Conclusions: Muscle strength is a better indicator of 5-year adverse clinical outcomes of mortality and low physical performance than muscle mass. Muscle strength was an independent predictor of low physical performance in 5 years. Geriatr Gerontol Int 2016; 16: 577–585. Keywords: mortality, muscle strength, older adults, physical performance, sarcopenia.

Introduction Sarcopenia is defined as a loss of muscle mass and strength with aging.1,2 Sarcopenia is associated with low physical performance, disability, hospitalisations and mortality.3,4 Muscle mass or strength (quality)-related indices have been considered as muscle indices for defining sarcopenia. Although conceptually, sarcopenia has been defined in terms of loss of muscle mass, consensus as to the most appropriate muscle index for the definition has not been achieved.2,5,6 Conventionally, cli-

Accepted for publication 20 March 2015. Correspondence: Dr Jae-Young Lim MD PhD, Department of Rehabilitation Medicine, Seoul National University College of Medicine, Seoul National University Bundang Hospital, 300 Gumi-dong, Bundang-gu, Seongnam-si, Gyeonggi-do, 463-707, Korea. Email: [email protected]

© 2015 Japan Geriatrics Society

nicians and researchers have used a definition of sarcopenia that employs only muscle mass as an index, and have defined this in different ways.5 However, the decline in muscle strength is faster than that of muscle mass in older people,7 and age-associated loss of muscle strength is not completely explained by the loss of muscle mass.8,9 Therefore, muscle mass, as a representative of muscle function, might be an index of lesser importance. From a clinical point of view, an appropriate indicator should be closely associated with clinical outcomes and should predict adverse outcomes, such as disability, and mortality. In this context, several studies have shown that muscle strength is more closely associated with poor health outcomes.6,10–16 Most of the evidence, however, has come from cross-sectional analyses.6,10,13 Although several longitudinal studies have shown that muscle strength is a better predictor of mortality14,15,17,18 and disability,12 these studies used diverse definitions of doi: 10.1111/ggi.12514

| 577

YH Kim et al.

muscle strength as the ratio of muscle size to strength (muscle quality),17 knee extensor strength12 or grip strength,14,15 and cut-off points as tertile,15 quartile12 or standard deviation,14 as with inconsistent definitions of muscle mass. Furthermore, the primary outcome of most studies was mortality only.14,15,17,18 Thus, the causal relationship between muscle strength and clinical outcomes except for mortality could not be determined clearly. Longitudinal studies including these multiple parameters at the same time are lacking to confirm causal relationships between muscle indices and adverse clinical outcomes. In a previous cross-sectional study, we found that muscle mass was not associated with physical performance in weak older adults, emphasising that muscle strength is a better index from a clinical perspective.19 The present study, which was part of a prospective observational cohort study with 5-year follow up, was carried out to investigate the relationship between muscle indices and 5-year clinical outcomes in a population of community-dwelling older people, and to determine independent predictors for adverse clinical outcomes after adjusting for potential confounders. We hypothesised that muscle strength would be a better predictor of clinical outcomes than would muscle mass.

Methods Study population The present analysis was carried out using data obtained during the Korean Longitudinal Study on Health and Aging, a population-based prospective cohort study of residents aged 65 years or older in Seongnam, Korea. A detailed description of the Korean Longitudinal Study on Health and Aging was published elsewhere.20 The baseline examination was carried out from September 2005 to September 2006. In total, 560 people (274 women, 286 men) for whom baseline data on anthropometric parameters, body composition and muscle strength measurement were available were included. Of the 560 participants, 41 died, and 355 (171 women, 184 men) completed 5-year follow-up examinations between May 2010 and March 2012. Causes of dropouts were incomplete data (n = 22) and declined to follow up (n = 142). The study protocol was approved by the institutional review board of Seoul National University Bundang Hospital. Informed consent was obtained from all participants.

Demographic and anthropometric data Trained and certified nurses recorded age, sex and comorbidities using standardized questionnaires. “Regular” exercise was defined as an exercise regimen of at least 30 min three times or more per week. Height, 578 |

waist circumference and bodyweight were measured to the nearest 0.1 cm and 0.1 kg, respectively. Body mass index was calculated as weight divided by height squared (kg/m2).

Muscle mass index Dual-energy X-ray absorptiometry (DXA) is an effective tool for research and clinical use to estimate muscle mass with minimal radiation.2 Total muscle mass (total lean mass) was measured by DXA (Lunar Corporation, Madison, WI, USA). Appendicular skeletal muscle mass (ASM) was calculated as the sum of the muscle mass of the four extremities. We used ASM divided by weight (ASM/Wt) and by height squared (ASM/Ht2) as muscle mass indices.2

Muscle strength index The isokinetic muscle strength of the knee extensors was measured using an isokinetic dynamometer (Biodex Isokinetic Tester; Biodex Medical Systems, Shirley, NY, USA) at an angular velocity of 60°/s. Participants carried out two sets of five repetitions, with a 30-s rest between sets, exerting maximum pressure on the arm of the isokinetic device. The concentric peak torque values (Nm) obtained from five torque-angle curves in each set were used. The torque values were normalized against bodyweight (Nm/kg) to reduce the dispersion resulting from a strong relationship between isokinetic muscle strength and the participant’s bodyweight.21 Grip strength was measured with a hand dynamometer (Hand Evaluation Kit; Sammons Preston, Bolingbrook, IL, USA) with participants seated, their elbow by their side and flexed to right angles, and a neutral wrist position. The measurement was carried out by a trained occupational therapist and the mean value of three measures was used in the analysis.

Evaluation of clinical outcomes The study outcomes were 5-year all-cause mortality and low physical performance in 5 years. All-cause mortality data over the 5-year period were obtained from the Ministry of Public Administration and Security national database. To evaluate physical performance, the Short Physical Performance Battery (SPPB) was administered at the 5-year follow-up visit. The SPPB is a standard functional outcome measure for research and clinical outcomes.22 A total possible score of 12 was created using the sum of four possible points on three tests: chair stands, gait speed and standing balance. Participants were first asked to balance in a standing position with their feet side by side, semi-tandem and fully tandem for 10 s each. Participants were next asked to walk a distance of 4 m as fast as possible. Finally, © 2015 Japan Geriatrics Society

Muscle strength and low performance

participants were asked to stand from a sitting position in a chair and return to the seated position five times as quickly as possible while keeping their arms folded across their chest. Low physical performance was defined as a SPPB score of less than 9.19,23

Covariates From among the broad range of geriatric evaluations carried out in the Korean Longitudinal Study on Health and Aging as covariates, we included fat mass,24 and evaluations of cognitive functioning25 and depressive symptoms20,26 as potential confounders. Total fat mass was measured using DXA. Percentage fat mass was calculated as total fat mass × 100/bodyweight. Cognition was assessed using the validated Mini-Mental State Examination (MMSE) of the Korean version of the Korean version of Consortium to Establish a Registry for Alzheimer’s Disease (CERAD-K) assessment packet.27 MMSE scores range from 0 to 30, with a higher score indicating better global cognition. The severity of depression was evaluated using the Korean version of the Geriatric Depression Scale (GDS).28 GDS scores range from 0 to 30, with a higher score indicating a more depressive state.

Statistical analysis The statistical significances of differences between participants with different clinical outcomes were assessed using the independent t-test or the Mann–Whitney U-test for continuous variables, and Pearson’s χ2-test or Fisher’s exact test for categorical variables, as appropriate. Spearman’s rho correlation for each variable was tested before the multivariate analyses. Significant variables from the univariate analyses were included as possible covariates in the final multivariate models. Adjusted Cox proportional hazard models were used to evaluate relationships between muscle indices and mortality. To identify independent predictors of low physical performance, a multivariate logistic regression analysis was carried out. All tests were two-sided, and a P-value less than 0.05 were taken as statistically significant. The PASW statistics version 18.0 (IBM Corporation, Chicago, IL, USA) was used for statistical calculations.

Results The participants who did not complete 5-year follow-up examination (n = 205) were older (P < 0.001), and had lower bodyweight (P = 0.048), higher waist-to-hip ratio (P = 0.003), lower total muscle mass (P = 0.035), lower ASM (P = 0.020), lower leg muscle strength (P < 0.001), lower grip strength (P < 0.001), lower MMSE (P < 0.001) and higher GDS (P < 0.001). Men © 2015 Japan Geriatrics Society

Figure 1 Correlation between total muscle mass and leg muscle strength in men and women.

showed a stronger linear correlation between baseline total muscle mass and leg muscle strength than women (R2 = 0.104 in women; R2 = 0.224 in men; Fig. 1) The stronger linear correlation in men was also observed between initial total muscle mass and grip strength (R2 = 0.105 in women; R2 = 0.204 in men).

Baseline factors associated with 5-year clinical outcomes Table 1 shows the characteristics of the participants alive and dead. Men and women who had died were significantly older, and had lower ASM, lower leg muscle strength, lower grip strength and lower MMSE scores than did those who lived to the 5-year follow up. In women, lower serum albumin, heart disease, lower fat mass and lower percentage body fat were significantly associated with 5-year mortality. In men, lower total muscle mass, lower normalized leg muscle strength and higher GDS scores were significantly associated with 5-year mortality. Characteristics of participants with high physical performance (SPPB ≥ 9) and with low physical performance (SPPB < 9) are presented in Table 2. Older age, lower leg muscle strength, lower normalized leg muscle strength and lower grip strength were associated with lower physical performance after 5 years in both sexes. Absence of regular exercise, stroke or parkinsonism, higher waist circumference, lower MMSE and higher GDS were associated with low physical performance only in women. | 579

YH Kim et al.

Table 1 Characteristics of the participants alive and dead

Age (years) No regular exercise Physical activity (h) Serum cholesterol (mg/dL) Serum albumin (g/dL) Serum triglycerides (mg/dL) Serum hsCRP (mg/dL) Comorbidities Hypertension Heart disease Stroke or parkinsonism Diabetes mellitus Cancer Chronic lung disease Arthritis Fracture Weight (kg) Body mass index (kg/m2) Waist circumference (cm) Waist-to-hip ratio Total muscle mass (kg) ASM (kg) ASM/Wt (%) ASM/Ht2 (kg/m2) Total fat mass (kg) Percentage fat mass (%) Leg muscle strength (Nm) Normalized leg muscle strength (Nm/kg) Grip strength (lbf) MMSE (points) GDS (points)

Women (n = 274) Alive (n = 259) Dead (n = 15)

P-value

Men (n = 286) Alive (n = 260)

Dead (n = 26)

P-value

73.14 ± 7.03 143 (55.2) 4.09 ± 1.74 209.98 ± 35.95 4.15 ± 0.21 143.71 ± 84.06 0.19 ± 0.38

79.39 ± 8.91 12 (80.0) 3.80 ± 1.57 211.60 ± 29.75 4.02 ± 0.27 138.67 ± 78.98 0.13 ± 0.19

0.002 0.074 0.761 0.864 0.021 0.948 0.559

73.66 ± 7.40 82 (31.5) 4.34 ± 1.90 194.73 ± 38.23 4.13 ± 0.22 137.99 ± 98.22 0.25 ± 0.84

79.35 ± 9.19 11 (42.3) 3.84 ± 1.64 186.00 ± 34.90 4.13 ± 0.24 118.35 ± 63.41 0.49 ± 0.96

0.002 0.264 0.086 0.252 0.692 0.270 0.144

121 (46.7) 43 (16.6) 14 (5.4) 43 (16.6) 24 (9.3) 23 (8.9) 143 (55.2) 49 (18.9) 61.03 ± 9.99 24.39 ± 3.11 86.78 ± 8.98 0.92 ± 0.09 40.60 ± 8.01 16.93 ± 3.97 27.70 ± 4.26 6.69 ± 1.02 17.69 ± 6.16 30.13 ± 8.81 64.04 ± 27.23 1.04 ± 0.38

9 (60.0) 6 (40.0) 2 (13.3) 1 (6.7) 1 (6.7) 2 (13.3) 7 (46.7) 3 (20.0) 58.97 ± 13.02 23.70 ± 3.71 87.58 ± 10.65 0.93 ± 0.07 40.24 ± 7.98 16.51 ± 4.00 28.09 ± 4.22 6.61 ± 0.99 15.45 ± 6.83 26.96 ± 8.29 52.49 ± 26.06 0.88 ± 0.39

0.323 0.034 0.216 0.478 1.000 0.636 0.507 1.000 0.128 0.122 0.652 0.781 0.057 0.013 0.440 0.256 0.002 0.013 0.025 0.089

111 (42.7) 34 (13.1) 11 (4.2) 45 (17.3) 22 (8.5) 21 (8.1) 51 (19.6) 32 (12.3) 65.56 ± 9.62 24.11 ± 3.17 87.75 ± 8.61 0.92 ± 0.07 47.14 ± 5.31 20.13 ± 2.75 30.91 ± 3.02 7.39 ± 0.85 15.65 ± 6.07 24.18 ± 6.88 79.45 ± 26.51 1.21 ± 0.37

15 (57.7) 4 (15.4) 3 (11.5) 9 (34.6) 4 (15.4) 3 (11.5) 6 (23.1) 4 (15.4) 63.36 ± 11.53 23.90 ± 3.74 89.57 ± 8.79 0.95 ± 0.06 44.91 ± 5.75 18.81 ± 3.02 29.83 ± 3.83 7.08 ± 0.87 15.67 ± 6.86 24.90 ± 7.43 61.33 ± 27.16 0.97 ± 0.42

0.142 0.761 0.124 0.061 0.274 0.467 0.674 0.550 0.283 0.757 0.329 0.107 0.044 0.021 0.090 0.074 0.985 0.515 0.002 0.003

42.09 ± 20.81 24.69 ± 3.62 10.37 ± 7.13

34.02 ± 18.52 21.93 ± 6.11 12.21 ± 7.04

0.012 0.006 0.574

55.15 ± 19.02 25.99 ± 2.72 8.63 ± 6.60

42.54 ± 15.35 23.50 ± 5.78 11.60 ± 7.23

0.001 0.007 0.036

Values are means ± SD or n (%), as appropriate. ASM, appendicular skeletal mass; ASM/Ht2, appendicular skeletal mass divided by height squared; ASM/Wt, appendicular skeletal mass divided by weight; hsCRP, high-sensitivity C-reactive protein; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale.

Independent predictors of 5-year clinical outcomes The Cox regression models used to identify mortality risks associated with muscle indices are shown in Table 3. With respect to the muscle strength indices, separate models were constructed for leg muscle strength and grip strength (models 1 and 2). Because strong correlations were found among total muscle mass, ASM and ASM/Ht2 (Spearman ρ = 0.879 to 0.980), only ASM/Ht2 was included in the final model as a muscle mass index. ASM/Wt was also not included in the model because of its moderate-to-strong correlation with fat indices, such as total fat mass and percentage fat mass (Spearman ρ = −0.626 to −0.875). All models were adjusted for age, percentage fat mass, MMSE and GDS. 580 |

In women, serum albumin was included in the model according to the result of univariate analysis. Neither muscle strength nor muscle mass indices were independent predictors of 5-year mortality in both sexes. In women, lower percentage fat mass and lower serum albumin level were independent predictors for mortality in 5 years. The logistic regression models used to identify independent predictors for low physical performance are shown in Table 3. As with the Cox regression models for mortality, separate models were constructed for each muscle strength index (models 1 and 2). Considering the strong correlations between muscle mass indices (Spearman ρ = 0.881 to 0.982), only ASM/Ht2 was included in the final model. ASM/Wt was also not © 2015 Japan Geriatrics Society

© 2015 Japan Geriatrics Society

73.76 ± 6.63 28 (68.3) 3.52 ± 1.53 210.22 ± 38.35 4.16 ± 0.21 137.27 ± 79.67 0.18 ± 0.48 17 (41.5) 4 (9.8) 4 (9.8) 9 (22.0) 2 (4.9) 2 (4.9) 25 (61.0) 11 (26.8) 57.64 ± 8.60 25.18 ± 2.87 87.84 ± 8.97 0.94 ± 0.12 34.08 ± 3.28 13.63 ± 1.78 23.87 ± 2.58 5.96 ± 0.60 20.77 ± 5.94 37.23 ± 6.09 43.15 ± 15.29 0.75 ± 0.25 28.02 ± 11.04 21.98 ± 3.83 15.32 ± 7.41

70.05 ± 4.68 58 (45.7) 4.07 ± 1.55 210.65 ± 36.83 4.16 ± 0.15 145.70 ± 86.59 0.19 ± 0.38 61 (47.3) 25 (19.4) 1 (0.8) 19 (14.7) 13 (10.1) 14 (10.9) 70 (54.3) 19 (14.7) 56.68 ± 7.52 24.42 ± 2.96 83.78 ± 9.21 0.90 ± 0.10 34.33 ± 3.67 13.92 ± 1.78 24.66 ± 2.40 6.00 ± 0.62 19.54 ± 5.35 35.73 ± 5.97 54.52 ± 15.32 0.97 ± 0.28 32.64 ± 12.82 25.04 ± 3.03 10.30 ± 3.58

0.514 0.154 0.012 0.277 0.527 0.363 0.451 0.077 0.491 0.154 0.030 0.102 0.700 0.504 0.074 0.699 0.212 0.164

Muscle strength: A better index of low physical performance than muscle mass in older adults.

The most appropriate muscle index for the definition of sarcopenia has not been agreed on. We aimed to investigate the associations of muscle mass and...
253KB Sizes 3 Downloads 10 Views